key: cord-1037053-mjwlzuxa authors: Straub, John E. title: A Multiscale Coarse-grained Model of the SARS-CoV-2 Virion date: 2021-01-26 journal: Biophys J DOI: 10.1016/j.bpj.2020.12.032 sha: de227637381db1da7f7d8925b162f9bccaf91e9e doc_id: 1037053 cord_uid: mjwlzuxa nan The global COVID-19 pandemic has led to intense activity to develop an effective vaccine. In addition, there have been efforts to develop therapeutics to treat symptoms of the disease, reduce the severity of infection, or prevent infection entirely. Inspired by the potential of structure-based therapeutic design, research has led to detailed molecular models of key protein structural components of the SARS-CoV-2 virion [1] . Motivation for this approach comes from the substantial success of past efforts to create effective therapeutics for HIV and influenza using rational design approaches [2, 3] . An important lesson of that past work is that the most effective rational design approaches account for multiple conformational states of the protein targets and the flexibility of pharmacophores [4, 5] . As such, the development of atomistic models of virion structures, that enable molecular dynamics simulations that can account for and assess the multiple conformational states defining the virion structural ensemble, are needed [6] . The SARS CoV-2 virion is composed of spike, membrane, nucleocapsid, and envelope proteins encoded by the virus's 30 kilobase (kb) genome. Significant progress has been made in modeling the spike proteins that mediate receptor recognition and cell membrane fusion. Other proteins play key roles in organizing RNA into ribonucleoprotein complexes, supporting the formation of ion channels in the virion membrane, and supporting viral budding. In order to elucidate the mechanism of viral replication and infectivity, dependent on the collective behavior of these many components, a holistic model of the virion is required. Cryo-electron microscopy (cryo-EM) and x-ray crystallographic studies have been successful in providing insight into the structure of key viral proteins. As substantial regions of these proteins are left unresolved, homology modeling and structure prediction methods have been used to complete structural models of component proteins [7] . While these experimental and computational methods provide insight into individual protein components, additional work is required to complete an integrated model of the complete SARS-CoV-2 virion. Complete models of virions are known and have been constructed for HIV-1 [8] . Artistic renderings of the molecular structure of the SARS-CoV-2 virion have been widely circulated in scientific writing and popular culture [9] . However, until recently we have lacked a detailed atomistic model that is based on state-of-the-art methods for translating the partial knowledge provided by experimental structural biology, homology modeling, and fold prediction into an integrated virion model. In "A Multiscale Coarse-grained Model of the SARS-CoV-2 Virion," the groups of Rommie Amaro and Greg Voth report the results of an impressive effort to use state-of-the-art computational methods to develop a complete model of the virion. Employing a "bottom-up" approach, they develop a coarse-grained (CG) model of the SARS-CoV-2 virion based on available experimental structural and atomistic simulation data. All-atom models of the key structural components of the virion were developed based on information from cryo-EM, X-ray cystallograhic structures, and homology models. Following the construction of models of the key virion components, all-atom simulations were used to inform the construction of coarse-grained models. A variety of coarse-graining methods pioneered in the Voth group were used to iteratively develop in a hierarchical manner a CG model for the full SARS-CoV-2 virion. The Essential-Dynamics Course-Graining (EDCG) approach was used to develop a coarse-grained model that preserves the principal modes of motion of the system [10] . Following the initial coarse-graining the intramolecular interactions within a protein unit were treated using an Elastic Network Model (ENM) [11] . In modeling non-bonded interactions, the CG potential energy function was represented as a linear combination of basis functions where the coefficients were determined using force-matching (FM) [12] or relative entropy minimization (REM) [13] in the Multi-Scale Coarse-Graining (MS-CG) approach [14] . Finally, to address large-scale conformational changes or bond cleavage, a multi-state Ultra-Coarse-Graining (UCG) approach may be effective [15] . Using the constructed virion model, CG dynamics were performed followed by an analysis of the structure and fluctuations. It was observed that the highest variance principal component was associated with splaying motions in the S1/S2 domain of the spike protein. That motion accounted for 51% of the observed fluctuations in the system. The next principal component due to rocking motions of the S1/S2 domain accounted for 12.5% of the fluctuation. Finally, a principal component associated with twisting of the S1/S2 domain accounted for 7% of the fluctuations. The authors conclude that "these correlated modes of motion are likely informative of how the virion collectively utilizes spike proteins to explore and detect receptors." This collaborative effort has established a valuable computational resource for the community of researchers focused on studies of SARS-CoV-2 that includes atomistic trajectory and experimental structural data deposited in the NSF Molecular Sciences Software Institute (MolSSI) as they become publicly available [16] . This resource will serve researchers as a platform to incorporate computational and experimental data. Given the success of past efforts to develop therapeutics for HIV and influenza using rational design approaches, we can be hopeful that similar approaches will be effective in the development of COVID-19 treatment and prevention strategies. A natural starting point will be multiscale modeling studies of the SARS-CoV-2 virion made possible by the creation of this resource. Those simulations stand to provide insight into structure-function relationships as well as the generation of structural ensembles for use in rational design approaches accounting for multiple conformational states of this complex molecular assembly. A Community Letter Regarding Sharing Biomolecular Simulation Data for COVID-19 Viracept (nelfinavir mesylate, AG1343): a potent, orally bioavailable inhibitor of HIV-1 protease Rational design of potent sialidase-based inhibitors of influenza virus replication Method for including the dynamic fluctuations of a protein in computer-aided drug design Developing a dynamic pharmacophore model for HIV-1 integrase Ensemble Docking in Drug Discovery Modeling of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) Proteins by Machine Learning and Physics-Based Refinement 3D molecular models of whole HIV-1 virions generated with cellPACK Inside the Coronavirus A Systematic Methodology for Defining Coarse-Grained Sites in Large Biomolecules Systematic Multiscale Parameterization of Heterogeneous Elastic Network Models of Proteins The multiscale coarse-graining method. I. A rigorous bridge between atomistic and coarse-grained models The relative entropy is fundamental to multiscale and inverse thermodynamic problems A Multiscale Coarse-Graining Method for Biomolecular Systems The Theory of Ultra-Coarse-Graining. 1. General Principles SARS-COV-2 Coarse Grained Viron Model DOI: 10.34974/q8ya-wh69 and Multiscale Models for the SARS-CoV-2 Virion